David Poehlman wrote:
I'm missing something. If I go from one page to another on a site, that seems to me to be history. What am I missing? If I go back, I should be able to retrace my steps.

Yep and that's all handled natively by typical browsers. Unless the back button won't work for some reason, the publisher gets this functionality for free.

If I come in from somewhere else, there should be no breadcrum brail unless I'm missing something.

A good breadcrumbs implementation has nothing to do with your browsing history, despite the name "breadcrumbs". I think it would be rather confusing to try and express surfing history as a chain of greater than relationships.

When you come in from another site or a distant cousin of the page elsewhere in the site hierarchy, you do not know the levels between the page you're now on and the site's homepage. Breadcrumbs immediately explain that genealogy and make each ancestor page immediately accessible.

For example, I arrive by search engine at the product page for a Hotpoint ZX500 washing machine at a white goods retailer's website. I might get a breadcrumb like this:

AppliancesAreUs > Catalog > Washing machines > Hotpoint ZX500

Perhaps Hotpoint ZX500 turns out to be not quite the washing machine for me. Now I can jump straight to see other products in the category I'm interested in with the washing machines link. I don't need to go to another page, trawl through the site map, and hunt around for the relevant category. And thank goodness for that, as it would make comparing washing machines much slower and I'd be tempted to go to another site.

If I want to know a sites structure, I shold be able to get it but
not be made to waide through it.

Nobody's "made to wade through it", since tools should make it easy to skip it and frequently do make it so.

A breadcrumb list is the genealogy of an individual webpage, not a site map of the entire website tree. A better name for the breadcrumbs design pattern would probably be "page ancestors" or "page genealogy".

--
Benjamin Hawkes-Lewis

Reply via email to